Max Robotics — Global Service Robot Directory + US Certification

About

Annett Stelzer is a robotics researcher specializing in autonomous navigation, visual perception, and biologically inspired algorithms for mobile and legged robots. Her work spans two complementary domains: vision-based navigation for complex terrain and scalable mapping strategies drawn from insect cognition. Stelzer's most influential contribution is her stereo-vision-based navigation system for the DLR Crawler, a six-legged walking robot capable of traversing unknown rough terrain using semi-global matching for depth perception combined with visual odometry — a paper that has accumulated 121 citations and stands as a landmark reference in legged robot navigation. Alongside this, she developed the Landmark-Tree map framework, a biologically inspired topological mapping approach designed to overcome the scalability limitations of traditional metric maps. This line of work, extended through the Trail-Map data structure and visual homing algorithms based on bearing angles to landmarks, demonstrates her sustained commitment to efficient, resource-conscious navigation suited to robots with limited computational capacity. Her leg proprioception-based odometry work further highlights her breadth, addressing motion estimation from the ground up in statically stable walking robots. Across her career, Stelzer has made meaningful contributions to making autonomous robots smarter, lighter in computation, and more adaptable to real-world environments.